In The Arena by TechArena - Studying the Earth from the Stars with NASA's Megan Damon

Episode Date: November 18, 2022

Tech Arena host Allyson Klein chats with NASA’s Megan Damon on her group’s study of global air pollutants and how supercomputing helps speed insight to global climate change and human welfare.

Welcome to the Tech Arena, featuring authentic discussions between tech's leading innovators and our host, Alison Klein. Now, let's step into the arena. Welcome to the tech arena. My name is Allison Klein. And to cap off our week of coverage at supercomputing, I'm so excited to have NASA's Megan Damon with us. Megan, welcome to the program. Thank you so much for having me, Allison. I'm happy to be here. Megan, why don't we just get started and introduce your background at NASA and what you're delivering today for the organization? Sure. Well, I've been working with NASA for over 15 years now. I came in originally to help out with some software engineering tasks, and that was for a group that was studying the chemistry of the

atmosphere. I also worked on hurricane mission support. I've also been working quite a bit with scientific workflows, which help researchers run high performance simulations in a more efficient manner. Right now I'm working with a group called the GOCF, and our focus is mostly on air pollution. And I'm also currently working with satellite data. But at supercomputing, I'm here to focus on forecast pollution. So your resume at NASA is a great indication of the broader purview of NASA. And I think that a lot of our listeners, rightfully so, associate NASA with space exploration. But the purview of the agency is much broader. Can you comment on that? Sure, absolutely. NASA certainly is very interested in space. But the key here is that NASA is interested in all planets, including our own.

And Earth is definitely the best planet for testing our instrumentation and technologies. It's certainly the most accessible, right? So the measurements that we're looking at in terms of Earth measurements and space measurements are really for all of humankind. And I wanted to point out that we're not just interested in activities that are happening in the United States, it's for the entire world. In fact, if you think about the NASA mission, the statement there is to drive advances in science, technology, aeronautics, and space exploration. And this is to enhance knowledge, education, innovation, economic vitality, and stewardship of Earth. That is quite a mouthful of charter. Yes, it is. And, you know, one of the

things that really struck me, I walked by the NASA booth at supercomputing, and you were giving a talk on air pollutants. And I was really blown away by what you guys were tracking across the globe. Can you tell me why air pollutants are important to study? And how they interact with that broader charter that you just talked about? Sure. Air pollutants are actually a massive health hazard. Millions of people die each year from air pollution. And in fact, it's the leading cause of premature death worldwide. So we're seeing about one out of eight premature deaths that are related to air pollution. And not only do we have to consider health impacts, but there are also negative impacts

that cost the world economy up to $8 trillion a year. And so we have satellites, not only NASA, but many other partners in the United States and international partners as well, have satellites that are helping to track these sources of air pollution. And what we need to do is we do need to better understand the impacts of these pollutants as well as devise smarter mitigation strategies. So when we couple these measurements that we obtain from satellites with theoretical models, this gives us air quality information. And not only does it give air quality information to institutions like NASA in the United States and beyond, but context on what types of disease or ailments are impacted by air pollution?

Sure. This certainly isn't my area of specialty, but we would consider respiratory diseases. That's an obvious one. Heart disease can often be linked to air pollution as well as strokes. Okay. So it's way beyond what a lot of people would think about in terms of respiratory illness. Yes. When we look at the science itself, and you went into a lot of detail on the science, what are the key things as climate evolves on the planet, What are the key things that you're looking at from a standpoint of the things that are causing air pollution? And is this something that's on

the rise? Sure. Air pollutants in greenhouse gases and the greenhouse gases trap heat into the atmosphere. And that's the cause of global warming. In particular, we're looking at things like carbon dioxide, methane, nitrous oxide, and water vapor. Now, certainly, some of these pollutants come from human activity. Some of them are, of course, natural. But we are looking at pollution that's being caused by transportation, industry, as well as agriculture causes as well. So there really is a plethora of systems to look at when concerning with air pollution, as well as global warming. Now, one of the things that I was struck by from your talk is how a local event could have global implications. Can you give some examples of where we've seen that and

why that's important to study? Sure, of course. Well, the key here is, is that air pollution doesn't know anything about borders or border crossing. Okay, there's no passports at the borders here. Saharan dust storms that start in Africa can blanket countries within Europe. Locally, if we think about the United States, fires that are coming from the US West Coast have been shown to cause bad air quality all the way to Washington, DC. Air pollution that's caused from industrial activities in eastern China travel all the way across the Pacific Ocean and those impact the West Coast of the United States. And then if we think about natural

sources of air pollution, the Honga Tonga eruption increased water vapor all the way through the upper layers of the atmosphere. So the climax of that volcanic eruption was in January 15, 2022. And so the water vapor, the heightened water vapor in the stratosphere is impacting global temperatures. So you just named off a number of different types of air pollutants. What are the key categories that you're tracking? I would say for the GOCF model, which is the model I'm here reporting on in supercomputing, we're specifically interested in ozone, particulate matter, which are small, fine particles. We also look at nitrogen dioxide as well and carbon monoxide. Now, all of the data that you're collecting is captured by satellites.

Is that correct? And not all of the data. So when we talk about Earth observations, we mostly think about satellites. However, there are a number of different ways to gather observations across the planet. We can make ground-based measurements. We can do balloon launches and catch you in situ observations. NASA and other agencies have aircraft missions which fly planes through specific pathways in order to gain readings that we need more information about. We can also measure observations from sea and air and ships.

So it's not just satellites, but satellites are the big cost, I would say. And what they create collectively are these massive visualizations that are giving you information in real time for what's in the air. Tell me about how you use that data. And I'm sure there's other things that come out of that data than just the visualizations. But how do you use the data and how does it inform NASA science as well as inform other agencies or other institutes around the world that are studying the planet? Sure. When we run our models and we make predictions, this assists us in understanding what is the best way to use the instrumentation on our satellites to improve the quality of data that the satellites are giving to us. And there's all sorts of examples of this beyond just satellite observations. There are ground monitoring networks that use our forecast data in order to better position their instruments to gain better readings as well.

Additionally, we provide air quality forecasts. We have a number of customers who are interested in air quality, whether regionally or globally. And so our forecasts are used by a number of different institutions. So it's sort of a symbiotic relationship, if you think about it, between air pollution forecasting using a theoretical model and the real-time satellite observations. We help them and they help us. Now, these visualizations obviously process a hefty amount of data to come up with global time-based views of what's happening around the planet. Can you give me a sense of the types of data sizes that your scientists are working with and the role of supercomputing

in helping with that? Sure, absolutely. So specifically, I'm going to talk about the GEOS composition forecasting model, again, which is our air pollution product at the center that I work out of. In our simulations, we represent the Earth's atmosphere as a three-dimensional grid. You can think of it as a really large cube. And that cube consists of 75 million grid points. And we also need to simulate both the troposphere and the stratosphere, which are two very important layers of the atmosphere for our studies. And so not only do we have a two-dimensional sense, which covers the latitude and longitude grid of the Earth, but we also have a three-dimensional aspect because the atmosphere begins at the ground and goes towards the top of the Earth's atmosphere. And in that sense,

we use 72 model layers. So we can think about this larger grid as a collection of smaller grids. And our spatial resolution in the latitude longitude dimension is 25 by 25 kilometers. So again, we need about 7 million data points in order to run this simulation. And the 75 million data points corresponds with spatial resolution, not time. Of course, we're marching through time in order to create predictions. Now, not only do we have this large grid, but we also have to transport more than 250 chemical species. So you can think about the grid I just described and then multiply that out by 750 because that's the number of chemical species we need. But in addition to chemical species, we need to know other properties about the atmosphere, such as winds, sunlight, temperature, humidity, things of that nature.

We call that meteorology. In addition to just thinking about the grid, we also have to take into consideration the calculation of how these chemicals are evolving through time. And we have more than 700 chemical reactions within our chemistry solver that we have to take into consideration to forecast air pollution. When you look at all of that from somebody who's been in the tech industry for a long time, that's a very large data set that you're calculating. And one question that I have for you, Megan, is, you know, do you have any perspective on how the industry can work with NASA to continue to support these types of calculations? And are there examples of how you work with the industry on this type of work? Well, NASA has a longstanding relationship with many international partners,

both academic and industry partners. So we've all been contributing to the larger mission in order to have a better understanding of our planet in general. To point to some specific examples, we've already seen massive improvements coming from the computing industry. This has allowed us to increase our model resolution. Another example would be using cloud services such as Amazon Web Services. And this has made it very easy for non-NASA parties to access air pollution data, as well as be able to manipulate that data. And then beyond just theoretical computations, one hot area in air pollution is the development of low-cost sensors.

And this is an area where the tech industry can fill a gap by providing high-quality and low-cost air quality surface information, both in real time and otherwise. And that can be an invaluable complement to satellite observations. So I'm sure that folks who are listening are really keen to take a look at some of this data and see the visualizations because they are really stunning. Yes, they are. Does NASA post those anywhere? And can we point somewhere where our audience can take a look? Absolutely. You can either Google the scientific visualization studio, or you can go directly to svs.gsfc.nasa.gov. And there you'll find not only a collection of high definition visualizations,

both in three dimensions and two dimensions, but there also will be information listed about the context in which those visualizations were produced. So you can watch them as well as learn. A couple more. Additionally, the Earth Information Center, I would just Google that because the link that I have written down here is probably too long to say to our audience. But the Earth Information Center that NASA offers gives a wonderful review of the Earth science endeavors that NASA is currently undertaking. And if you're interested in our forecasting products from the GOCF model, you can go to a website called Fluid. That's fluid.nccs.nasa.gov. Well, Megan, I think I'll be doing some Google searching myself after this interview to check it all out. One final question for you. If folks want to follow along on the earth science endeavors on social, where would you point them?

I would point them to NASA Earth on Twitter, as well as NASA Expeditions on Instagram. Fantastic. Thank you so much for being with us today and taking time out of your busy supercomputing schedule. Thank you so much, Allison. It was a pleasure to be here. Thanks for joining the Tech Arena. Subscribe and engage at our website, thetecharena.net. All content is copyright by the Tech Arena.